Don't Retract Pack

Where Should Babies Sleep at Night?

By Peter Fleming

Several strong recommendations have been made recently by different groups concerning the safest or most appropriate sleeping arrangements for infants in Western society. The US Consumer Product Safety Commission, basing its information on reports of child deaths in adult beds, has advised parents against taking their baby into bed with them. Unfortunately, the study upon which this recommendation is based did not include a control group; thus, while individual cases were described in which accidental asphyxia was adduced from the circumstances of death, no population data were collected to allow an assessment of actual risk.

In the UK, the Sudden Unexpected Deaths in Infancy study (SUDI), carried out as part of the Confidential Enquiry into Stillbirths and Deaths in Infancy (CESDI), was a population-based case-control study of all sudden unexpected deaths of infants from 7 to 365 days of age, in five regions of England. (1,2,3,4) It was the largest such study yet conducted and included all unexpected infant deaths occurring over a three-year period from a population of 470,000 births. The families of infants who died were contacted as soon as possible after the death (usually within three or four days), and a nurse-researcher interviewed the parents to collect detailed information on the family background and medical, social, economic, and environmental factors; in all, the database included more than 600 fields. Particular emphasis was given to the precise sequence of events and a detailed description of the place in which the infant was sleeping when put down for the last sleep, and when found dead.

For each infant who died, four live infants were selected, matched for age, locality, and date, and precisely similar information was collected from the parents of these infants. Detailed information was collected on the sleeping conditions in a "reference" sleep, which occurred in the 24 hours before the interview, to ensure that parents remembered the details.

The study included information on 450 infants who had died suddenly and unexpectedly and 1,800 matched "control" infants. For 325 of the infants who had died, no explanation was found for the death, which was therefore certified as due to sudden infant death syndrome. The CESDI SUDI study thus allows us to examine in great detail the conditions in which infants were sleeping and to compare the sleeping arrangements in which infants died with those in which infants of the same age, in the same society, were sleeping at the same time. From this comparison it is possible to identify conditions that differed between the live infants and those who died.

The study offers a unique opportunity to examine the evidence as to whether sharing a bed with a parent was a factor contributing to the risk of death. In order to carry out such a comparison, it is important to consider factors that may affect the risk of bedsharing, for example, whether the parent had consumed alcohol or taken sleep-inducing drugs. A previous study in New Zealand had indicated that sharing a bed with a parent who smoked might be hazardous, so this factor also needed to be taken into account. (5,6)

The CESDI study showed that for infants who shared a room with a parent, the risk of SIDS was approximately half that for infants who slept alone. In other words, putting a baby to sleep in a separate room (rather than the room in which the parents slept) doubled the risk of SIDS. For parents who smoked, had been drinking alcohol, or were excessively tired (less than 4 hours uninterrupted sleep in the previous 24 hours), sharing a bed with their baby increased the risk of SIDS for babies less than four months of age. For babies over this age, or for parents who did not smoke and had not been drinking or taking sleep-inducing drugs, there was no evidence that bedsharing increased the risk of SIDS.

Sleeping with a baby on a sofa or armchair was found to be extremely hazardous, increasing the risk of SIDS by a factor of 25. (7,8)

From the information given by the parents of the live babies, it is clear that around half of them brought their baby into bed with them at times, and at times they would fall asleep while the baby was there. Careful analysis of the CESDI data showed that at least half of all parents bedshared with their baby at some time.

The CESDI study confirmed previous studies that showed that putting a baby to sleep on its stomach, under excessive bedding, or under bedding that could ride up over the baby's head, or using a pillow or soft bedding in the crib, would increase the risk of SIDS. Avoiding such factors is thus an appropriate way to reduce the risk of SIDS and is estimated to have saved around 100,000 infants' lives worldwide. (Estimate based upon recorded falls in SIDS rates and infant mortality in the US, Europe, Australasia, and the UK.)

No one would suggest that because sleeping in a crib can be hazardous under certain conditions, no baby should sleep in a crib. By analogy, therefore, it is equally illogical to suggest that because under certain circumstances bedsharing can be hazardous, parents should not bedshare with their baby. Given the near universality of the practice of bedsharing at some stage, it is far more logical to identify the conditions under which bedsharing is hazardous and to give parents information on how to avoid them.
From the CESDI study data, the most logical recommendations would be:

  • Always put your baby to sleep on his back, not on his side or front.
  • Do sleep with your baby in the same room; putting her in a separate room doubles the risk of SIDS.
  • If you find it easier, particularly if you are breastfeeding, do bring your baby into bed to feed.
  • If you smoke, have been drinking alcohol, have taken drugs or medicines that may make you sleepy, or are excessively tired, do not bring your baby into bed with you to sleep; put him back into the crib after he feeds.
  • Do not sleep with your baby on a sofa, armchair, waterbed, or very soft mattress.
  • If you wish to sleep with your baby in your bed, make sure that the bedding cannot cover her head, and keep her away from the pillow.

These recommendations take into account the known risk factors and if implemented will significantly reduce the risk of SIDS. It is important to note that in the CESDI study, although around 8 percent of infants shared a bed with nonsmoking parents, only six infants (2 percent) died in bed with a nonsmoking parent.

If bedsharing with nonsmoking parents were hazardous-if, for example, it increased the risk of SIDS by 50 percent-we would have expected at least 40 such deaths; and even if it were neither a risk nor protective, we might have expected 26 such deaths. The very small number of deaths in bed with nonsmoking parents in this study suggests that bedsharing with nonsmoking parents is not a significant risk factor for SIDS.

Rather than issuing broad statements, not based upon good evidence, to suggest that parents should not bedshare with their babies, I suggest that giving them accurate information, based upon careful studies of healthy babies as well as babies who have died, will allow parents to make safe and appropriate choices.


1. P. S. Blair, P. J. Fleming, I. J. Smith, et al., "Babies Sleeping with Parents: Case-control Study of Factors Influencing the Risk of the Sudden Infant Death Syndrome," BMJ 319 (1999): 1457-1462.
2. P. J. Fleming, P. S. Blair, C. Bacon, P. J. Berry, eds., Sudden Unexpected Death in Infancy: The CESDI SUDI Studies 1993-1996 (London: The Stationery Office, 2000).
3. K. Pollard, P. J. Fleming, J. Young, et al., "Night time Non-nutritive Sucking in Infants Aged 1 to 5 months: Relationship with Infant State, Breastfeeding, and Bed- versus Room-sharing," Early Human Development 56 (1999): 185-204.
4. P. Blair, P. J. Fleming, D. Bensley, I. Smith, et al., "Smoking and Sudden Infant Death Syndrome: Results of 1993-5 Case-control Study for the Confidential Enquiry into Stillbirths and Deaths in Infancy," BMJ 313 (1996): 195-198.
5. R. Scragg, E. A. Mitchell, B. Taylor, "Bed sharing, Smoking and Alcohol in the Sudden Infant Death Syndrome," BMJ 311 (1995): 1269-1272.
6. E. A. Mitchell, P. G. Tuohy, J. M. Brunt, et al., "Risk Factors for Sudden Infant Death Following the Prevention Campaign in New Zealand: A Prospective Study," Pediatrics 100 (1997): 835-840.
7. See Note 1.
8. See Note 2.

Peter Fleming, CBE, PhD, MBChB, FRCP, FRCP (C), FRCPCH, is a professor of infant health and developmental physiology at the University of Bristol and a pediatrician at the Royal Hospital for Children, Bristol, United Kingdom.


1 comment:

  1. I would be interested in knowing the exact percentage ofSIDS-based deaths when co-sleeping compared to sleeping alone when the alcohol/drug factor is removed. Is there any data on this?

    The above noted that this increased the risk (For parents who smoked, had been drinking alcohol, or were excessively tired (less than 4 hours uninterrupted sleep in the previous 24 hours), sharing a bed with their baby increased the risk of SIDS for babies less than four months of age) but did not indicate by what percentage the rate increased. Also, it was unclear to me whether or not these instances were included in the overall number of bed-sharing based SIDs deaths or not? Thank you for the information!